Global Journal of Human Social Science, G: Linguistics and Education, Volume 23 Issue 8

Figure 2: Measurement model Figure 2 shows that two items (REL3 and REL4) from reliability, one item from responsiveness (RES1) and one item from empathy (EMP4) were deleted as dimensions of SERVEQUAL. Similarly, four items from the assessment (ASS2, ASS3, ASS4 and ASS5) were also deleted because their loadings fell below the acceptable threshold to enhance their reliabilities. All other items were retained because their loadings were in line with the minimum threshold. In the same vein, the HTMT criterion and cross-loading were used to assess the discriminant validity. Table 1 HTMT criterion E-learning Satisfaction Servequal E-learning Satisfaction 0.786 Servequal 0.817 0.879 Note: Satisfaction stands for students' satisfaction, while serv-equal stands for teaching quality. Table 1 shows the Heterotrait-monotrait (HTMT) ratio for all the latent variables. The HTMT compares the values to a predefined threshold. The HTMT ratio of correlation, according to Henseler et al. (2015), is a superior ratio than the Fornell-Lacker and the cross- loading criterion due to its higher specificity and sensitivity rate. As decision criteria, a value close to 1 indicates a lack of discriminant validity. Thus, some scholars (Gold et al., 2001; Kline, 2011) recommend a threshold of 0.85 and 0.90. For this study, the maximum threshold of 0.90 was used to decide the HTMT. Therefore, the values of 0.879 for this study are below the HTMT0.90 and indicate no multicollinearity among the variables, as Gold et al. (2001) argued. b) Structural model evaluation This section presents the results of the structural equation model testing the study's hypotheses. The p-value at a 5% level of significance was used to accept or reject the hypotheses, as shown in Table 2. Volume XXIII Issue VIII Version I 86 ( ) Global Journal of Human Social Science - Year 2023 G © 2023 Global Journals Effect of Teaching Quality on Students' Satisfaction in Nigerian Tertiary Institutions: The Moderating Role of E-Learning Amid COVID-19 Recovery